Motion Aware Self-Supervision for Generic Event Boundary Detection

TL;DR

This paper introduces a simple self-supervised method augmented with a differentiable motion feature learning module for generic event boundary detection in videos, demonstrating effectiveness on challenging datasets without explicit motion pretext tasks.

Contribution

It presents a straightforward self-supervised approach with a novel motion feature learning module, simplifying the GEBD task and outperforming complex existing methods.

Findings

Effective on Kinetics-GEBD and TAPOS datasets

Learns motion features without explicit motion pretext tasks

Outperforms state-of-the-art self-supervised methods

Abstract

The task of Generic Event Boundary Detection (GEBD) aims to detect moments in videos that are naturally perceived by humans as generic and taxonomy-free event boundaries. Modeling the dynamically evolving temporal and spatial changes in a video makes GEBD a difficult problem to solve. Existing approaches involve very complex and sophisticated pipelines in terms of architectural design choices, hence creating a need for more straightforward and simplified approaches. In this work, we address this issue by revisiting a simple and effective self-supervised method and augment it with a differentiable motion feature learning module to tackle the spatial and temporal diversities in the GEBD task. We perform extensive experiments on the challenging Kinetics-GEBD and TAPOS datasets to demonstrate the efficacy of the proposed approach compared to the other self-supervised state-of-the-art methods. We also show that this simple self-supervised approach learns motion features without any explicit motion-specific pretext task.